The Depressing Effect of Agricultural Institutions on the Prewar Japanese Economy

"The Depressing Effect of Agricultural Institutions on the Prewar Japanese Economy" by Fumio Hayashi and Edward C. Prescott (Journal of Political Economy, 2008, vol. 116, no. 4).

This paper attempts to address the question of why the "Japanese miracle" did not occur before the war. It starts out with the premise that "for much of the prewar period of 1885-1940, Japan's real GNP per worker remained a third of that of the United States."

The idea here is that, essentially, since first sons of farmers were basically required to take up farming, the number of farmers stayed the same. This constancy in the number of farmers preserved inefficient, small scale farming techniques, and locked up too much of the labor force for the country to fully industrialize and urbanize.

Some interesting implications are: 1) If this is the case, then perhaps it was nothing more than the breaking of this agricultural tradition which allowed Japan's economy to flourish in the decades after the war; 2) Perhaps, had Japan "solved" this "problem" sooner, it would have had a greater industrial output before and during the war, possibly prolonging or even changing the outcome of the war.

My personal assessment is that, were Japan's industrial output to have been 25-50% higher from about 1935-1945, there would have been more American casualties, the war would have cost far more for the US in terms of money, and would have left Japan in possession of some of the easternmost of its Pacific acquisitions. It is not unrealistic to speculate that the West Coast of the US might have been attacked more heavily under such a scenario.

Specifically, using a two-factor neoclassical growth model and assuming that labor constraints were a major factor in constraining Japan's growth, the paper's model predicts that prewar output would have been 32% higher in Japan without the labor constraint.

A major problem with this article is that it presupposes the postwar output was somehow the "real" or equilibrium output and that prewar Japan was simply failing to achieve it. It is also very possible that Japan was overachieving during that post-war period, however, which would invalidate this entire analysis. They do address this issue, but somewhat inconclusively, in my opinion. Other ways of stating this problem are: "maybe the average productivity of labor in Japan was unusually high during the miracle" or "maybe the traditional one-third to capital, two-thirds to labor split" was strongly and exceptionally violated during the miracle.

Another major problem with this article is it is difficult to separate the confounding factor of technological change from the anthropological issue of first sons staying on farms. For example, the tractor was not introduced into japan until 1955. Perhaps the impact of the tractor in Japan was particularly large, and this accounts for some or all of the pre-to-post-War disparity.

Inefficient Unemployment Dynamics under Asymmetric Information

"Inefficient Unemployment Dynamics under Asymmetric Information" by Veronica Guerrieri (Journal of Political Economy, 2008, vol. 116, no. 4.).

This article sets out to discuss what it considers to be a novel externality generated by information asymmetry between job seekers and employers: companies fail to understand the impacts their previous contract offerings have on current negotiations.

Guerrieri accepts the notion that competitive searching for jobs is good. For example, if job seekers can see posted salaries, they will direct their searches more efficiently, and this, ultimately causes the surplus created by the negotiated job contract to be more efficiently distributed. However, she still believes that the asymmetry of information might cause problems.

Basically, the problems work like this: When workers have private information about their own options, they can capture part of this in negotiating a contract. Firms who post contracts at time t+1 affect the options that a worker faces at time t. However, this information is not captured by other firms, causing them to act on incomplete information.

If there is a certain level of unemployment today which is above the steady-state level, then the number of workers who could meet and negotiate with an employer today is higher than it is likely to be tomorrow. Therefore, the average information distortion is also higher today. If job creation tomorrow could be reduced, then there would be less value to continuing one's search, which would reduce the wages a worker would accept, increase employment, and ultimately increase job creation.

On a fundamental level, Guerrieri is going down this path in an attempt to argue that some kind of central planner responsible for intervening in the labor market could improve it. The connections and arguments above seem to me to be tenuous at best.

This is a problem that plagues this type of literature more than just this specific essay, but the raw quantity of unrealistic assumptions required to make these arguments renders them difficult, at best, to apply. For example, in this case, both employers and employees are risk neutral, and employees face no cost of job searching, whereas employers face a cost of posting jobs.

More than anything else, though, this article rests upon one assumption that I don't know to be right. It rests on the assumption that after a shock to the economy, the desire must be to return to the "equilibrium" level of job creation as soon as possible. While this is a nice idea, a shock such as the bursting of a bubble may leave the economy in a situation where the equilibrium rate of job creation and destruction is precisely what is not known (to planners or mere forecasters).

Learned Journals

Sometimes, I read articles in various scholarly journals, including, but not limited to: Foreign Affairs, Foreign Policy, The Journal of International Security Affairs, Political Economy, and Policy Review. In order to record what I read for myself, and to share it with others, I intend to start blogging my responses to these articles, here.